Prepare training data with reddit comments (tree-structure)
- Download the dataset you wish to train.
- Open chatbot_database.py and change the directory to where your own dataset is located
- Modify the functions accordingly to your own conditions
- Run the program and create database to store your data
Create training data with reddit comments (tree-structure)
- Open chatbot_database_training_data.py
- Change directory to the created database
- Run the program to create the training data
- Git clone the project from Daniel Kukiela, https://github.com/daniel-kukiela/nmt-chatbot
- Modify the project accordingly to the setup that suits you best
- Replace the training data with your own training data
- Run the project and train your chatbot